ABSTRACT
Objectives:
To evaluate the role of 18F-fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) for the diagnosis of anterior mediastinal masses.
Methods:
The oncological 18F-FDG PET/CT images of 41 patients (17 women, 24 men; age: 16-83 years, mean age: 50.5±19.5 years) who attended the nuclear medicine department between November 2016 and September 2017 were retrospectively evaluated for the metabolic characterization of their anterior mediastinal masses.
Results:
Based on our results, the lesions of 4 patients were benign [maximum standard uptake value (SUVmax) <3] and that of 2 patients were non-tumoral (i.e., tuberculosis and sarcoidosis). The mean dimensions and the SUVmax levels of the malignant lesions were 6.4±3.7 cm and 11.9±9.6, respectively. The pathological results for the malign tumors were thymus tumors (n=8), lymphoma (n=8), lung cancer (n=11), carcinoid metastasis (n=2), thyroid carcinoma (n=2), germ cell carcinoma (n=1), schwannoma (n=1), and sarcoma (n=1). The degree of 18F-FDG accumulation could precisely identify the malign and benign tumors.
Conclusion:
Thus, contrary to the known causes, it is possible that anterior mediastinal masses originate from structures other than the anterior mediastinal structures. In this study, the lymphoma and lung carcinoma pathology were more frequent than thymic lesions.
Introduction
Several causes of anterior mediastinal masses, either benign or malignant, have been reported. In fact, it has been reported that anatomic structures in the anterior mediastinum either enlarge or become malignant or metastasize from another tumor. Several studies have also evaluated thymus enlargement as a differential diagnosis of anterior mediastinal masses (1). In addition, numerous cases have been reported related to the origin of anterior mediastinal masses in the literature (1). The other reasons of development of anterior mediastinal mass are benign enlargement of anatomic structures involving or invading the mediastinum (2,3).
It has been previously documented that the 18F-fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) is an accurate modality in the staging and restaging of anterior mediastinal tumors. In fact, a recent study suggested that the diagnostic power of 18F-FDG PET/CT for anterior mediastinal mass is high, but its negative predictive value is higher (4). Another small series of study determined the cut-off level for the determination of malignancy in anterior mediastinal mass (5). The present study aimed to evaluate the pathological outcomes of adult patients presenting with anterior mediastinal mass in conjunction with the 18F-FDG PET/CT findings.
Materials and Methods
Patients
The following were the patient inclusion criteria: age of 16-85 years and presenting with anterior mediastinal mass without histopathological diagnosis.
Informed consents were obtained from the patients for conducting PET/CT examinations.
The following were the patient exclusion criteria: pregnancy and lactation, age <16 or >85 years, presenting with another malign tumor elsewhere, and those with contraindication for PET/CT examinations.
The study was approved by the Local Ethics Committee and conducted according to the revised Helsinki Declaration, 2010. Mersin University Rectorate Clinical Research Ethics Committee (date: 05/10/2017, no: 2017/285).
The PET/CT images of 41 patients (17 women, 24 men, age: 16-83 years, mean age: 50.5±19.5 years) who were referred to the nuclear medicine department with the diagnosis of anterior mediastinal mass by a previous CT examination conducted between November 2016 and September 2017 were obtained. The data were retrospectively evaluated by 2 experienced nuclear medicine physicians without any knowledge of the final diagnosis of the patients. The PET/CT study was not performed for patients with anamnesis of pregnancy and lactation and for those with contraindication for the examination. In addition, we did not prefer pediatric patients with interfering problems that involved the anterior mediastinum frequently (such as physiological thymus activity) and elder patients (age: >85 years) because they probably could not be operated.
PET/CT Examinations
The patients were prepared for the examination by ensuring at least 6 h of fasting and decreased physical activities since at least 24 h prior to the examination. The patients were first injected with the radiopharmaceutical agent [mean 370 MBq (10 mCi), according to the body weight] via the venous line 60 min before the imaging. Imaging was performed by using a PET/CT scanner (discovery PET/CT 610; GE, US) with a low-dose CT scan (130 kV, 50 mAs, 1.5 pitch, 5-mm thickness, 70-cm field of view) for attenuation correction without intravenous contrast administration via oral contrast administration from the skull base to the upper thigh region with the acquisition time of 3 min/bed position and the matrix size of 256x256. Attenuation-corrected PET images were then reconstructed by using an iterative reconstruction algorithm, VUE point HD with 3 iterations and 32 subsets.
Diagnostic Criteria
The images were evaluated with respect to the metabolic characteristics of the anterior mediastinal lesions [maximum standard uptake value (SUVmax)] levels obtained from the workstation (Mac iOs, Osirix MD programme). The SUVmax levels were retrieved by the circular region of interest covering the most active portions of the lesions in addition to the CT characteristics of the lesions and the dissemination to other structures (metastatic dissemination, lymphadenopathies elsewhere in the body, and other possible malignant primary sites). The anterior mediastinal mass lesions were determined to be benign in case of a single site with low uptake of 18F-FDG (SUVmax <3).
Interventions and Histopathological Analysis
Surgical procedures were decided with reference to the PET/CT imaging and suspected malignancy. The types of surgical procedures conducted (such as thoracotomy, minithoracotomy sternotomy procedures, or biopsy) for each patient are summarized in Table 1. The final pathological outcomes obtained from the specimens of surgery including hematoxylene and eosine or immunohistochemistry (in case it was necessary) staining procedures were analyzed by an experienced pathology physician, and the results of the PET/CT and pathology were compared.
Statistical Analysis
The statistical analysis was performed by using a package program (MedCalc®v10.3.0). The receiver operating curve (ROC) analysis was performed in order to determine the power of the SUVmax parameter to differentiate between the benign and malignant lesions.
Results
Of the 41 study participants, 37 underwent different surgeries based on their imaging findings (Table 1). The pathological results including those of 2 patients with the diagnosis of granulomatous diseases (Figure 1) are listed in Table 1. A total of 4 patients were considered to be benign based on their PET/CT imaging, these patients did not undergo any surgical procedure and were also out of the follow-up program (Figure 2). In this series, 10 patients were diagnosed with lymphoma, while 2 were diagnosed with neuroendocrine tumor metastases with a relatively low 18F-FDG uptake. One of the patients had immature teratoma with a significantly high metabolic activity (Figure 3). The patient diagnosed with differentiated thyroid carcinoma also showed high 18F-FDG affinity; however, those with medullary thyroid carcinoma showed a relatively low uptake. Unexpected results, such as lung carcinoma, were recorded in the study group as well (Figure 4).
The distribution of the pathological results and the SUVmax levels of the lesions are summarized in Table 1. The 18F-FDG avid lesions outside of the anterior mediastinum of the patients are listed under Table 1. Thirteen patients underwent follow-up 18F-FDG PET/CT imaging (mean 7.4±5.2 month), while 3 patients died within 1 month of PET/CT examination during the disease course. The progression of the disease was noted in 4 patients, with partial response in 2 and complete metabolic response in 5. One patient was diagnosed with interfering infection and 2 with recurrent mediastinal lesion.
The cut-off SUVmax level during the determination of malignant and benign tumors was 6.04 as per the ROC curves, and, with this cut-off value, the sensitivity and specificity of the diagnostic modality was 74% and 80%, respectively (Figure 5). The SUVmax level was found to be significantly (p=0.05) successful in the determination of malignant lesions, while this level in patients with thymoma was 3-5.95. The SUVmax levels of thymic carcinoma was significantly higher than those of thymoma lesions (3-19,39).
Discussion
The benign anterior mediastinal masses may be benign metastasizing leiomyoma, inflammatory endobronchial pseudotumor, physiological thymus activity, or rebound thymus activity (1). The benign metastasizing leiomyoma is a rare tumor of the middle-age women occurring years after hysterectomy (6). Inflammatory pseudotumor is a benign lesion of children or young adults that may be associated with trauma, paraneoplastic syndrome, or inflammatory reactions (7). It is therefore important to discriminate the thymus uptake or thymic rebound from the pathological uptake in PET/CT among young people and children (1). Unfortunately, patients with benign lesions in this case series were also out of the follow-up and did not want to undergo an operation.
Although some pitfalls and false-positive results are associated with the 18F-FDG uptake of the surrounding tissues (8), the 18F-FDG PET/CT remains the most important modality in the preoperative evaluation of anterior mediastinal lesions. In this case series, only a limited number of patients showed false-positive results, including sarcoidosis (Figure 1) and tuberculosis. Granulomatous infections frequently interfere with malignant tumor metastasis or primary tumors, especially for PET/CT examinations in endemic countries and provides false-positive results. However, the ratio of false-positive results of patients with granulomatous diseases was found to be in an acceptable range in this study.
The most common tumors of the anterior mediastinum are thymic tumors “thymoma” in adults, and previous studies have demonstrated a close correlation between the World Health Organization and Mosaka stage of the thymic tumors and SUVmax levels (9). This study group included only 3 patients with thymoma, which is rarer than expected, while the SUVmax levels of these patients were in an acceptable range. The thymic carcinoma group showed higher SUVmax levels, as expected, than the thymoma group. CT or magnetic resonance showed extension and invasion into the adjacent mediastinal structures of the thymic tumors (9). The 18F-FDG PET/CT may differentiate the subgroup of thymic epithelial tumors such as thymoma, thymic carcinoma, and carcinoid tumors and accurately stage these tumors (10). Complete surgical removal of the thymic tumors with the involved adjacent structures is hence the popular treatment modality (11). A special subgroup of thymic tumors is the cystic thymus tumor that is characterized by multiple cysts (12). The 18F-FDG PET/CT imaging may demonstrate an increased uptake in the cystic thymomas related to the septum or margins of the tumor (12). The SUVmax cut-off level for malignant and benign thymomas have not been determined, although the previous reports accepted the 4.5-6.3 levels (13). The SUVmax cut-off value for the determination of the benign and malignant lesions by 18F-FDG PET/CT was 6.04 in this study by the ROC analysis. We noted that the diagnostic sensitivity and specificity of the test was acceptably high with this determined cut-off level.
Another problem with the anterior mediastinum is the physiological uptake related to thymus and a phenomenon called the “thymic rebound” that occurs especially in the pediatric age. Thymic rebound refers to the thymus regression during and enlargement after the completion of chemotherapy (1). A previous related study concluded that the SUVmax cut-off level of 3.1 may accurately differentiate thymic rebound from lymphoma recurrence (1). Another study about benign and malignant anterior mediastinal mass indicated the cut-off level of 3 (5). Other malignant tumors of anterior mediastinum include lymphoproliferative diseases such as lymphoma, neuroendocrine tumors, and mesenchymal tumors. The most important change in the patients’ management was observed in the lymphoma group since the biopsy sites were altered due to the PET results. In a case series on patients with lymphoma, it was suggested that mild 18F-FDG accumulating lesions in the anterior mediastinal region may indicate benign lesions like thymic hyperplasia even in lymphoma patients (14). However, in the case of the presence of diagnostic criteria indicating malignancy in lymphoma patients, secondary malignant tumors of the anterior mediastinum have been indicated in previous case reports (15).
Study Limitations
The study limitations include the retrospective design that limits the patient selection. In addition, relatively small number of patients could be included since the main subject of the study considered specific patient population.
Conclusion
This study thus demonstrated that adult patients, especially, may have several unexpected other primary malignancies, especially lung cancer (30%). No study has so far reported the pathological outcomes of anterior mediastinal mass in comparison with the 18F-FDG PET/CT imaging results. We demonstrated that 18F-FDG PET/CT is an essential imaging modality for the characterization of anterior mediastinal mass. This modality may change the patients’ management by determines other possible biopsy sites.